Re: transform as a presentation attribute [was: Re: [whatwg] SVG cloning elements from HTML5]

On Tue, Jun 24, 2014 at 12:48 PM, Juergen Roethig
<roethig@dhbw-karlsruhe.de> wrote:
> Tab Atkins Jr. wrote:
>> On Tue, Jun 24, 2014 at 9:00 AM, Juergen Roethig
>> <roethig@dhbw-karlsruhe.de> wrote:
>>>
>>> I may make 1 (one) coordinate system unit be of size 1px,
>>> 42px, 4711px, or even just 0.0815px, dependent on the viewbox, scaling of
>>> containing elements, and resolution of the output device. And now I get
>>> told
>>> that I should use a unit "pixel" for that? Incredible!
>>
>>
>> You may be surprised to learn that 1px does not necessarily equal 1
>> device pixel even in normal CSS.  It's a screen-independent size based
>> on an angular measure.
>
> No, I will neither be surprised about nor learn that at all, since I already
> know that a "device pixel" and a "pixel" might be different. Nevertheless,
> on desktop browsers, they very often have the same size. But you might be
> surprised to learn that especially on mobile devices (IOS, Android,
> whatever), "pixel" and "device pixel" are usually of different sizes - if
> those are of the same size, it's just an exception. That's why I regard a
> "pixel" usually as a "virtual device pixel" ;-) But as far as I see, I never
> wrote "device pixel" in my previous message, I just wrote "pixel", and the
> abbreviation "px" should be a synonym for just that. But I am quite sure
> that you will neither be surprised about nor learn that at all, since you
> already know that ...

So, all "transform" does is transform the coordinate space that those
units work in.  This is identical between CSS and SVG; some SVG
presentation attributes just allow px lengths to be specified without
a unit, while the CSS properties require units.  So yes, your
coordinate system unit might be 5px wide *in the untransformed
viewport's coordinate space*, but in the local space it's still just
1px, because user coordinates *are* px units when untransformed.
Viewbox is a scale/translate transform applied after the other
transforms.

> ... but I am surprised to learn, just now, about the accuracy, carefulness,
> and reliability when making new "standards" in the W3C - so far the "living
> standard" of HTML was the best example for that (and yes, I know, the term
> "living standard" for HTML is not originating from the W3C), but other
> standards seem to adopt those superiour principles as well (irony intended,
> or in that case you might even call it sarcasm).

Cool story bro.

~TJ

Received on Tuesday, 24 June 2014 19:55:59 UTC